CHAPTER 1

SEMICONDUCTOR FUNDAMENTALS

[Link] Structure
In our effort to understand the operation of semiconductors, a fundamental question we might
ask is “What is the internal structure of an atom?” Please understand that it is nonsensical to
ask what an atom might “look like” because its components are all smaller than the shortest
wavelengths of light that humans can see. Instead, we simply need a model to explain its
observed behaviour. Perhaps the most prolific model in the popular imagination is the
planetary model shown in Figure 1.1. In this model, the core, or nucleus, is drawn at the
centre and contains positively charged protons and non-charged neutrons. Revolving around
this core are negatively charged electrons, each following a nice, regular, planar path much
like a planet around the sun.

Figure 1.1. Planetary atomic Model

Most of the mass of any given atom is from the protons and neutrons. Protons and neutrons
have similar masses, about 1.67E−24 grams each. The mass of an electron is roughly 2000
times smaller. The radius of a proton is approximately 0.87E−15 meters and the mean
distance to the nearest electron is about 5.3E−11 meters. This means that this electron is
about 60,000 times farther away from the proton than the size of said proton.

1|Page
Another model used to represent an atom is the Bohr model, named after Danish physicist
Niels Bohr. An example is shown in Figure 1.2. It is important to understand that the Bohr
model is an energy description of the atom, not an attempt to mimic its physical appearance
or structure. The nucleus is placed at the centre. It is surrounded by concentric rings that
represent the electron shells. The higher the number, the larger the ring and the greater the
energy level. If an electron were to move from a higher level to a lower level, the energy
difference is radiated out. This could be in the form of heat or light. This is a point worth
remembering. For example, this transition is what makes light emitting diodes (LEDs)
function. The inverse is also possible, namely that by absorbing energy, an electron can move
into a higher orbital. This is an equally powerful concept, as we shall soon see.

the atomic number is indicated at the nucleus. Again, please do not imagine this representing
individual electrons orbiting the nucleus in lanes. This is an energy level depiction.

Figure 1.2. Generic Bohr model

Using the Bohr model we can create diagrams to represent individual elements. For
example, copper has an atomic number of 29 meaning that it has 29 protons and 29
electrons. The electron shell configuration is 2-8-18-1. That is, the first three shells are
completely filled and there is a single electron in the fourth shell. This single outer electron
is only loosely bound and thus makes copper a very good conductor. The Bohr model for
copper would simply show four rings, the first three being filled and with a single electron
in the fourth ring. Figure 1.3 shows the Bohr model of an atom of Silicon, atomic number
14, with an electron shell configuration of 2-8-4. In this version, the individual electrons
are drawn in each shell and

2|Page
 Figure 1.3. Bohr model of Silicon

Often, it is useful to simplify this model further by omitting the filled inner shells. Also,
the atomic number is replaced by the number of electrons in the outermost, or valence,
shell. This is shown in Figure 1.4. The valence shell is particularly important as it gives
insight into the general behaviour of the material.

Figure 1.4. Simplified Bohr model of Silicon

As an alternative, sometimes we will “straighten out” the Bohr model so that it simply
shows the energy levels graphically as lines or bands, and without counting specific
electrons. This is depicted in Figure 1.5.

3|Page
 Figure 1.5. Energy Level Diagram

[Link]
We used silicon in the preceding example on purpose. The fact that it has a half-filled valence
shell with four electrons puts it in a special place. As it's neither a great conductor nor a
superior insulator. With some attention to detail, it will become a semiconductor. Silicon is
not the only material that can be used for semiconductors. In fact, many of the earliest
semiconductors were made from germanium and currently we make semiconductors from
other materials. Silicon, however, remains the source of most semiconductors today. It is
possible for pure silicon to be arranged in a mono crystalline structure. That is, all of the
silicon atoms align in a very specific, well-ordered manner, without any voids or breaks
in the pattern. As silicon has only four electrons in its valence shell, four more electrons
would be needed to obtain stability (i.e., eight electrons in the outer shell). In the crystal,
any given atom of silicon effectively “shares” an electron from its four closest neighbours
through a covalent bond (meaning “with or among the valence”). Each atom is tightly bound
to its neighbours. This is illustrated in Figure 1.6 using simplified Bohr models. Note the
colour coding that indicates the sharing.

4|Page
 Figure 1.6. Silicon Crystal covalent bonding

Remember, this is an energy diagram. We are not trying to indicate that individual valence
electrons are zipping between atoms in figure eight patterns. Indeed, this diagram is drawn
flat whereas a real crystal is not a simple sheet, but is three dimensional with varying thickness.
While it would be a practical impossibility to draw a highly realistic representation of atoms
in the crystal, given a few liberties we can draw something that at least comes a little closer
to reality. We start by representing each silicon atom as a ball and the covalent bond as a
connecting tube. Recalling that each atom must be bound to four others in a regular, equal
pattern, we come up with the drawing of Figure 1.7. Notice that the overall structure is
essentially that of a cube. Further, at the centre of each face of the cube there exists an atom
of silicon. Therefore, we say that the crystal structure is face cantered cubic.

Figure 1.7. Representation of silicon crystal structure

5|Page
An interesting thing happens in the crystal when we examine the energy levels. With a single
atom we would expect to see an energy diagram like that of Figure 1.5. That is, discrete,
permissible steps. Within a crystal, though, each atom is affected by those around it. This
causes slight changes in the energy levels. Taken as a whole, all of these individual variations
cause the discrete levels to blur into broader bands. If we were to examine the valence and
conduction energy levels, instead of discrete, thin lines we'd see the thicker bands as
illustrated in Figure 1.8. These bands still represent permissible electron energy levels, it's
just that now there is a continuum rather than a discrete level. There will still be non-
permissible or forbidden zones between these regions. A forbidden zone is referred to
as a band gap.

Associated with this idea is the concept of the Fermi level, named after physicist Enrico Fermi.
Basically, the Fermi level is the energy level in a given material at which there is a 50%
probability that it is filled with electrons. In other words, levels below this value tend to
be filled with electrons and levels above tend to be empty. If the Fermi level lies within a
band, the material will be good a conductor. On the other hand, if the Fermi level lies between
two widely separated bands, the material will be a good insulator. If the Fermi level is between
bands that are relatively close, the material is a semiconductor.

Figure 1.8. Energy diagram for an intrinsic semiconductor

6|Page
Figure 1.8 shows the energy bands for an intrinsic semiconductor, such as an ideal silicon
crystal. The term intrinsic simply means that there are no impurities in the crystal.
Between the valence band and conduction band is an impermissible or forbidden region. This
is a band gap. In practical terms you can think of the band gap as the amount of energy that
needs to be applied to an electron in order to move it from the valence band to the conduction
band. The value of the band gap will depend on a variety of factors, the precise material being
used for the semiconductor is of particular importance. Without any external energy applied
(i.e., isolated and at absolute zero), the crystal lattice is stable and there is no electron
movement through the crystal. As we add thermal energy, it is possible for valence electrons
to jump up to the conduction band. At this point, the electron can “wander” through the crystal
in the manner depicted in Figure 1.9

Figure 1.9. Electron movement in a crystal

Here is how this happens: Because the thermal energy causes the electron to jump to the
higher energy level of the conduction band, it leaves behind a “hole”, that is, a place devoid
of an electron. Now that the hole exists, it provides a place for another electron to “fall into”.
The higher the temperature, the greater the number of freed electrons and the greater the
number of corresponding holes. We now have thermally-induced electron movement. We can
also look at this from the opposing perspective, namely that we have an equal magnitude but
opposite direction “hole flow”. If you find this idea hard to grasp, simply look at Figure 1.10.

7|Page
Each horizontal bar contains four dots representing electrons. In the topmost bar there is an
empty space (a hole) to the extreme left. When the leftmost electron moves into this hole it
fills it in a process called electron-hole recombination, which of course, sounds much more
impressive than it really is. The result is the second bar. We repeat this process of moving an
electron right to left as we traverse down the diagram. Eventually we end up with the four
electrons packed together toward the left. Finally, instead of focusing on the dots, focus
instead on the negative space (the empty white bit). Moving from top to bottom, the hole
moves left to right, in the opposing direction. Just as we think of the movement of electrons
as a movement of negative charge, then the movement of holes can be thought of as a
movement of positive charge. We can say that the electron is the carrier of negative charge
while the hole is the carrier of positive charge. Before moving on to the next section, it is
important to remember that in an intrinsic (pure) semiconductor, the number of
thermally produced electrons and holes will be equal. Also, even at room temperature the
total number will also be quite small compared to the number of electrons in the crystal.

Figure 1.10. Electron versus hole flow

1.3. Doped Material

By themselves, intrinsic semiconductors are not of particular use. They are neither good
conductors nor insulators, and their conduction is largely dependent on temperature.
We can alter the properties of the material by introducing foreign substances or impurities
into the crystal. These impurities are also known as dopants. A crystal with an added
dopant is referred to as an extrinsic semiconductor or doped material.

8|Page
There are two different types of semiconductors possible. One is called N-type material, and
the other, P-type material. Unsurprisingly, the N stands for Negative and the P stands for
(you guessed it) Positive. N-type material is created by adding pentavalent impurities, that
is, a dopant with five electrons in its outer shell. Examples include phosphorus, arsenic and
antimony. In contrast, P-type material is created by adding a trivalent impurity, one with
three electrons in its outer shell. Possible trivalent impurities include boron, gallium and
indium.

1.3.1. N-Type Material

Figure 1.11 shows a model of a silicon crystal with a pentavalent impurity at its centre.
Compared to an ordinary silicon atom that would have four electrons in its outer shell, the
pentavalent impurity creates an extra, or donor, electron. Thus, the crystal has a net negative
charge and is referred to as N-type material. The energy level of the donor electrons is just
below the bottom of the conduction band. In other words, the difference between the donor
level and the bottom of the conduction band is much, much smaller than the band gap itself.
Therefore it is relatively easy for these donor electrons to jump into the conduction band,
becoming free ionized electrons and leaving behind ionized holes.

Figure 1.11. Crystal with added pentavalent impurity (N-type)

9|Page
Compared to the undoped intrinsic crystal, the doped extrinsic crystal exhibits a relatively
high number of free electrons. This enhances the conductivity of the material, and the greater
the doping level, the greater the enhancement. Earlier it was mentioned that both electrons
and holes can serve as charge carriers. Because the number of free electrons is significantly
larger than the number of holes in N-type material, electrons in N-type material are
referred to as the majority charge carrier (or more simply, the majority carrier) while
holes are referred to as the minority charge carrier (or minority carrier). The extra
electrons add to the number of filled energy states and, being of higher energy than the valence
electrons, push the Fermi level to a higher value. Remember, the Fermi level represents the
point where 50% of states would be filled, so if we add states above this, then the new 50%
point must be higher than the former level. This is illustrated in Figure 1.14.

Figure 1.12. Energy band diagram of N-type semiconductor

1.3.2. P-Type Material

In a similar manner, if we introduce a trivalent impurity, our crystal model now features a
hole; a location where an electron is lacking. For this reason, trivalent impurities are
sometimes called acceptors. The resulting crystal model is illustrated in Figure 1.13

10 | P a g e
 Figure 1.13. Crystal with added trivalent impurity (P-type)
The resulting situation is essentially the reverse of that of the N-type material. Figure 1.14.
shows the energy band diagram for our new P-type material. In this case, the Fermi level has
been pushed down, closer to the valence band.

Figure 1.14. Energy band diagram of P-type semiconductor

In P-type material, holes out number free electrons. Consequently, holes are referred to as the
majority carrier in P material while electrons take on the role of minority charge carrier. As
with N-type material, the greater the amount of trivalent impurity added, the greater the
overall effect. By itself, a doped crystal can be used to create a resistor. The resistivity of the
material is a function of the doping level.

11 | P a g e